U.S. patent number 4,580,105 [Application Number 06/695,050] was granted by the patent office on 1986-04-01 for automatic reduction of intermodulation products in high power linear amplifiers.
This patent grant is currently assigned to AT&T Bell Laboratories. Invention is credited to Robert E. Myer.
United States Patent |
4,580,105 |
Myer |
April 1, 1986 |
Automatic reduction of intermodulation products in high power
linear amplifiers
Abstract
A distortion simulating pilot is injected at the input of an
amplifier which uses feed forward distortion correction. The
magnitude of the pilot signal in the amplifier output is used to
control a decreasing step size circuit algorithm for adjusting the
gain and phase of the feed forward distortion signal to eliminate
substantially the pilot signal and the distortion introduced by the
amplifier.
Inventors: |
Myer; Robert E. (Denville,
NJ) |
Assignee: |
AT&T Bell Laboratories
(Murray Hill, NJ)
|
Family
ID: |
24791353 |
Appl.
No.: |
06/695,050 |
Filed: |
January 25, 1985 |
Current U.S.
Class: |
330/149;
330/52 |
Current CPC
Class: |
H03F
1/3235 (20130101); H03F 2201/3212 (20130101) |
Current International
Class: |
H03F
1/32 (20060101); H03F 001/26 () |
Field of
Search: |
;330/52,149,151 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
"Adaptive Linearization of Power Amplifiers in Digital Radio
Systems", By A. A. M. Saleh et al., Bell System Technical Journal,
vol. 62, No. 4, Apr. 1983, pp. 1019-1033. .
"Correction of Non-Linearity in Low and Medium Power Transmitters
and Transposers" by C. Cluniat, Review Technique Thomson-CSF, vol.
10, No. 2, Jun. 1978, pp. 1-46..
|
Primary Examiner: Mullins; James B.
Assistant Examiner: Wan; G.
Attorney, Agent or Firm: Phelan; Charles Scott Nimtz; Robert
O. Roberts; Patrick E.
Claims
What is claimed is:
1. An automatic control system for reducing the distortion produced
by a power amplifier said control system comprising
means for inserting a pilot signal into the path of an input signal
before delivery to said power amplifier,
means for extracting a sample of the output signal from said power
amplifier,
means for delivering a signal representing the amplitude of said
pilot signal present in said output signal from said extracted
sample, and
means responsive to said signal representing the amplitude of said
pilot signal present in said output signal for automatically
controlling the attenuation and phase of a distortion component
which is derived from said output signal so that when said
controlled distortion component is coupled with said output signal,
substantially all distortion, including said pilot signal, is
canceled leaving a substantially pure amplified signal.
2. The automatic control system of claim 1 wherein said means
responsive to said signal representing the amplitude of said pilot
signal present in said output signal for automatically controlling
the attenuation and phase of the distortion component comprises
means for comparing said amplitude of said pilot signal in two
successive time intervals to determine whether said amplitude has
changed and for producing a signal representative of said change,
and
means responsive to said signal representative of said change for
adjusting said attenuation and said phase of said distortion
component.
3. The automatic control system of claim 2 further comprising means
for preventing said system from malfunctioning when said pilot
signal is lost temporarily.
4. The automatic control system of claim 2 wherein said means
responsive to said signal representing the amplitude of said pilot
signal present in said output signal further for automatically
controlling the attenuation and phase of the distortion component
comprises means for controlling the step size of said signal
representative of said change when said automatic control system is
first turned on.
5. A method for automatically controlling the distortion produced
by a power amplifier comprising the steps of
inserting a pilot signal into the input signal before delivery to
said power amplifier,
extracting a sample of the output signal from said power
amplifier,
delivering a signal representing the amplitude of said pilot signal
present in said output signal from said extracted sample, and in
response to said signal representing the amplitude said pilot
signal present in said output signal
automatically controlling the attenuation and phase of the
distortion component which is derived from said output signal so
that when said controlled distortion component is coupled with said
output signal, substantially all distortion, including said pilot
signal, is canceled leaving a substantially pure amplified
signal.
6. The method of claim 5 wherein step for automatically controlling
the attenuation and phase of the distortion component in response
to said signal representing the amplitude of said pilot signal
comprises the steps of
comparing said amplitude of said pilot signal in two successive
time intervals to determine whether said amplitude has changed and
for producing a signal representative of said change, and
adjusting said attenuation and said phase of said distortion
component in response to said signal representative of said
change.
7. The method of claim 6 further comprising the step of preventing
the amplifier system from malfunctioning when said pilot signal is
lost temporarily.
8. The method of claim 6 wherein said step of automatically
controlling the attenuation and phase of the distortion component
further comprises the step of controlling the step size of said
signal representative of said change when said automatic control
system is first turned on.
Description
TECHNICAL FIELD
This invention relates to high power linear amplifiers and, in
particular, to an automatic control system using a pilot tone for
the reduction of distortion produced by high power linear
amplifiers.
BACKGROUND OF THE INVENTION
All linear amplifiers distort the signal at some power level. This
distortion produces intermodulation products when multiple signals
are present. Intermodulation products are undesirable because they
cause interference and crosstalk. Standards have been set to limit
the level of these unwanted signals in transmitters. To meet these
standards, methods of reducing distortion have been developed.
The most common method is called feedback. Feedback works well at
low frequencies but it becomes a problem at ultra high frequencies.
At these frequencies two basic methods are generally used. They are
predistortion and feed forward.
Predistortion involves producing a distortion similar to the
distortion being generated by the linear amplifier and adding it at
the input in the correct gain, phase and delay to produce
cancellation of the distortion at the output of the linear
amplifier. This method requires matching the distortion
characteristics of two amplifiers and hence limits the amount of
correction that can be obtained.
The feed forward method does not have this limitation because it
separates out the distortion generated in the linear amplifier
itself, and then adds it back with gain, phase, and delay adjusted
for maximum cancellation. The amount of distortion reduction
available using feed forward is limited by the accuracy of the gain
and phase adjustments. Continuous precision trimming of these
adjustments is necessary to achieve and maintain the maximum
distortion reduction.
SUMMARY OF THE INVENTION
A distortion simulating pilot is injected at the input of an
amplifier which uses feed forward distortion correction. The
magnitude of the pilot signal in the amplifier output is used to
control a decreasing step size circuit algorithm for adjusting the
amplitude and phase of the feed forward distortion signal to
eliminate substantially the pilot signal and the distortion
introduced by the amplifier.
A search limiter circuit is provided to prevent the circuit from
locking up if the pilot signal is temporarily lost.
A course control circuit is included for taking a large step size
when the system is first turned on.
The present invention provides 360 degrees phase adjustment and up
to about twenty decibels of gain adjustment.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 shows the prior art;
FIG. 2 shows the circuit in block diagram useful to effect the
present invention;
FIGS. 3, 4 and 6 show the control circuit of FIG. 2 in various
levels of detail; and
FIG. 5 is a timing diagram for the control circuit.
DETAILED DESCRIPTION
Referring to FIG. 1, there is shown a prior art circuit in block
diagram form of a feed forward system. Splitter circuit 12 causes
the input signal on lead 11 to be duplicated: one part is sent to
power amplifier 14 and the other to canceler circuit 18 via path
15. The output from power amplifier 14 includes a distortion
component caused by the amplifying step. A small portion of the
output signal from power amplifier 14 is obtained from directional
coupler 16 and sent to cancellation circuit 18. The gain, phase and
delay of the input signal via lead 15 is adjusted so that the input
signal is canceled when combined with the output signal from power
amplifier 14 in cancellation circuit 18 to derive a pure distortion
component on lead 19. When this distortion component on lead 19,
after being adjusted for gain and phase, is combined with the
signal from the power amplifier output received via path 17 at
directional coupler 10, a clean signal is delivered at the output
from directional coupler 10 because the distortion component is
canceled. A problem with this method, however, is that the amount
of cancellation depends on the precision of the gain and phase
adjustments.
Referring to FIG. 2, there is shown in block diagram form the
preferred embodiment of the present invention. A test signal, or
pilot signal, is inserted into the path of the input signal via
coupler 30 where they are mixed before delivery to the input of
power amplifier 24. The level, that is amplitude, of the pilot
signal is adjusted to be equal to the level of the distortion
components being generated in power amplifier 24. This is typically
about thirty decibels below the desired signal level. The amplitude
and delay of the clean input signal via path 25 are adjusted to
equal the amplitude and delay of the distortion output sample; the
phase, however, is adjusted to be exactly opposite. The input
signal via path 25 and the input signal component from coupler 26
cancel each other at summing pad 28 leaving the distortion present
on output lead 29 from summing pad 28. The gain, phase and delay
adjustments at this point are not critical because the adjustments
are necessary only to reduce the clean signal down to the level of
the distortion to obtain good results.
The distortion components, however, require precise adjustment of
gain and phase to produce maximum cancellation of distortion when
added back into the delayed output. This adjustment is made by the
automatic control circuit 32. The reference for control circuit 32
is the pilot signal which is detected using narrow band receiver
34. Samples of the output from coupler 20, which represents the
point where distortion is being canceled, are obtained from coupler
36 and delivered to narrow band receiver 34. The amplitude of the
pilot is detected and used by control circuit 32 to determine the
precise gain and phase adjustments for circuit 40 which are
necessary to produce the best cancellation of both the pilot and
the distortion introduced by power amplifier 24.
Referring to FIG. 3, there is shown a block diagram of control
circuit 32 in FIG. 2. The output from level detector 310 is
proportional to the logarithm (log) of the pilot amplitude. Under
control of switch timing control circuit 350, switch 370 causes
current sources in circuit 360 to be connected to voltage storage
buffers 380. The output signals from buffer 380 via leads 33 and
35, respectively, control the gain and phase adjustments of circuit
40 for substantially eliminating the distortion components.
Each time switch 370 is closed, a small adjustment is made to
circuit 40. The direction of the change is detected by sensor 330
which controls the polarity of the current sources in circuit 360.
If the pilot amplitude is reduced by the adjustment, the polarity
remains the same for the next adjustment. If the pilot amplitude is
increased by the adjustment, the polarity of the current source is
reversed for the next adjustment. This process reduces the pilot
amplitude to a minimum. The size of each change is controlled by
the amount of current being delivered to voltage storage buffers
380. This current is made proportional to the amplitude of the
pilot so that as the ideal settings are approached the adjustments
become finer.
Referring to FIG. 4, there is shown further detail for the entire
electronic gain and phase controller comprising: narrow band pilot
receiver and level detector 34 of FIG. 2; automatic control circuit
32 of FIG. 2; attenuator and phase circuit 40 which comprises
ninety degree splitter 410, two biphase attenuators 420 and 430,
and summing pad 440; and, amplifier 42. Such an arrangement allows
360 degrees of phase adjustment and about twenty decibels of gain
change.
Switch timing control circuit 350 of FIG. 3 comprises decade
sequencer 352 as shown in FIG. 4. This sequencer 352 controls four
switches: switch 370 which comprises switches S.sub.1 and S.sub.2 ;
and, switches S.sub.3 and S.sub.4 which are part of direction of
change sensor 330. Sequencer 352 also controls two gates G.sub.1
and G.sub.2. FIG. 5 is a timing diagram showing the sequence in
which these switches and gates are controlled.
At time t.sub.0 when S.sub.3 is closed, capacitor C.sub.3 begins
charging to the pilot level. At time t.sub.1, S.sub.3 opens,
S.sub.1 is closed, capacitor C.sub.1 begins charging, and the phase
and gain changes of the distortion signal are effected via the
biphase attenuator 430 in circuit 40. Then at time t.sub.2, switch
S.sub.1 opens, switch S.sub.4 is closed, and capacitor C.sub.4
begins charging to the pilot level. At time t.sub.3, switch S.sub.4
opens, and comparator 332 determines whether the pilot level
changed up or down. At time t.sub.4, gate G.sub.1 is enabled, flip
flop 342 is toggled, that is, the current is reversed if the output
from comparator 332 indicates the pilot level has increased. At
time t.sub.5, the enabling signal to gate G.sub.1 is removed.
The aforesaid steps are repeated at times t.sub.5 through t.sub.9,
for operating switches S.sub.3, S.sub.4, S.sub.2, gate G.sub.2,
flip flop 344, for charging and discharging capacitors C.sub.3,
C.sub.4, and C.sub.2 and for adjusting biphase attenuator 420.
Thereafter at time t.sub.0, the enabling signal to gate G.sub.2 is
removed and the entire sequence recycled repeatedly. As the ideal
settings of gain and phase are approached, the size of each
adjustment becomes smaller.
FIG. 6 shows greater detail than FIG. 4 for control circuit 32 of
FIG. 2. The operation of each the components is known to one
skilled in the art and is not repeated here. Two refinements,
however, have been included in FIG. 6. One is a search limiter 60
which prevents a lock up condition from occurring if the pilot is
temporarily lost. The search limiter contains four comparators
which sense the positive and negative voltage levels at the buffer
outputs 33, 35. When these voltages approach their limit the
comparators reverse the current sources. This prevents the voltage
output from limiting and locking. The other refinement is the
addition of a course control circuit 66, or large step size search
circuit, for faster initial adjustments, when the circuit is turned
on.
* * * * *